Sorry, but can you please take a minute out of your busy day to link specifically to that article that you have "posted many times".? I did a search and only came up with things saying axial preload was important for stiffness, but did not see anything about periodic errors.
If your claim was that axial preload is all that is needed, I disagree.? Before this fix, I tried many different kinds of axial preload such as:
squeeze the bearing blocks by different amounts
purposely set the blocks too far apart and bias the mount east heavy - this still preloads the bearing
purposely leave the mounting bolt a little loose and squeeze the blocks together with a c-clamp plus rubber padding
None of these fixed the problem.
Another thing - I don't have 76 second "spikes".? I have a very smooth sinusoid with only the tiniest amount of second harmonic.? A bearing ball rolling over a piece of dirt will give spikes and throw out all kinds of harmonics as seen the Fourier transform.? Therefore I buy into the theory that it is due to a misaligned bearing only "leaning on" one ball at a time and the sinusoidal axial modulators are due to the load smoothly transitioning from one ball to the next.
In any case, I don't care that much anymore.? The local witch doctor cured my ailment.
